efi_64.c 22.2 KB
Newer Older
H
Huang, Ying 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 * x86_64 specific EFI support functions
 * Based on Extensible Firmware Interface Specification version 1.0
 *
 * Copyright (C) 2005-2008 Intel Co.
 *	Fenghua Yu <fenghua.yu@intel.com>
 *	Bibo Mao <bibo.mao@intel.com>
 *	Chandramouli Narayanan <mouli@linux.intel.com>
 *	Huang Ying <ying.huang@intel.com>
 *
 * Code to convert EFI to E820 map has been implemented in elilo bootloader
 * based on a EFI patch by Edgar Hucek. Based on the E820 map, the page table
 * is setup appropriately for EFI runtime code.
 * - mouli 06/14/2007.
 *
 */

18 19
#define pr_fmt(fmt) "efi: " fmt

H
Huang, Ying 已提交
20 21 22 23 24 25 26
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/types.h>
#include <linux/spinlock.h>
#include <linux/bootmem.h>
#include <linux/ioport.h>
27
#include <linux/init.h>
28
#include <linux/mc146818rtc.h>
H
Huang, Ying 已提交
29 30 31 32
#include <linux/efi.h>
#include <linux/uaccess.h>
#include <linux/io.h>
#include <linux/reboot.h>
33
#include <linux/slab.h>
34
#include <linux/ucs2_string.h>
H
Huang, Ying 已提交
35 36 37

#include <asm/setup.h>
#include <asm/page.h>
38
#include <asm/e820/api.h>
H
Huang, Ying 已提交
39 40 41 42
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
#include <asm/proto.h>
#include <asm/efi.h>
43
#include <asm/cacheflush.h>
44
#include <asm/fixmap.h>
45
#include <asm/realmode.h>
46
#include <asm/time.h>
47
#include <asm/pgalloc.h>
H
Huang, Ying 已提交
48

49
/*
50
 * We allocate runtime services regions top-down, starting from -4G, i.e.
51 52
 * 0xffff_ffff_0000_0000 and limit EFI VA mapping space to 64G.
 */
53
static u64 efi_va = EFI_VA_START;
54

55
struct efi_scratch efi_scratch;
56

57
static void __init early_code_mapping_set_exec(int executable)
H
Huang, Ying 已提交
58 59 60
{
	efi_memory_desc_t *md;

61 62 63
	if (!(__supported_pte_mask & _PAGE_NX))
		return;

64
	/* Make EFI service code area executable */
65
	for_each_efi_memory_desc(md) {
66 67
		if (md->type == EFI_RUNTIME_SERVICES_CODE ||
		    md->type == EFI_BOOT_SERVICES_CODE)
68
			efi_set_executable(md, executable);
H
Huang, Ying 已提交
69 70 71
	}
}

72
pgd_t * __init efi_call_phys_prolog(void)
H
Huang, Ying 已提交
73
{
74 75 76 77
	unsigned long vaddr, addr_pgd, addr_p4d, addr_pud;
	pgd_t *save_pgd, *pgd_k, *pgd_efi;
	p4d_t *p4d, *p4d_k, *p4d_efi;
	pud_t *pud;
78

79
	int pgd;
80
	int n_pgds, i, j;
H
Huang, Ying 已提交
81

82 83 84 85 86
	if (!efi_enabled(EFI_OLD_MEMMAP)) {
		save_pgd = (pgd_t *)read_cr3();
		write_cr3((unsigned long)efi_scratch.efi_pgt);
		goto out;
	}
87

88
	early_code_mapping_set_exec(1);
89 90

	n_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT), PGDIR_SIZE);
91
	save_pgd = kmalloc_array(n_pgds, sizeof(*save_pgd), GFP_KERNEL);
92

93 94 95 96 97 98 99 100
	/*
	 * Build 1:1 identity mapping for efi=old_map usage. Note that
	 * PAGE_OFFSET is PGDIR_SIZE aligned when KASLR is disabled, while
	 * it is PUD_SIZE ALIGNED with KASLR enabled. So for a given physical
	 * address X, the pud_index(X) != pud_index(__va(X)), we can only copy
	 * PUD entry of __va(X) to fill in pud entry of X to build 1:1 mapping.
	 * This means here we can only reuse the PMD tables of the direct mapping.
	 */
101
	for (pgd = 0; pgd < n_pgds; pgd++) {
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
		addr_pgd = (unsigned long)(pgd * PGDIR_SIZE);
		vaddr = (unsigned long)__va(pgd * PGDIR_SIZE);
		pgd_efi = pgd_offset_k(addr_pgd);
		save_pgd[pgd] = *pgd_efi;

		p4d = p4d_alloc(&init_mm, pgd_efi, addr_pgd);
		if (!p4d) {
			pr_err("Failed to allocate p4d table!\n");
			goto out;
		}

		for (i = 0; i < PTRS_PER_P4D; i++) {
			addr_p4d = addr_pgd + i * P4D_SIZE;
			p4d_efi = p4d + p4d_index(addr_p4d);

			pud = pud_alloc(&init_mm, p4d_efi, addr_p4d);
			if (!pud) {
				pr_err("Failed to allocate pud table!\n");
				goto out;
			}

			for (j = 0; j < PTRS_PER_PUD; j++) {
				addr_pud = addr_p4d + j * PUD_SIZE;

				if (addr_pud > (max_pfn << PAGE_SHIFT))
					break;

				vaddr = (unsigned long)__va(addr_pud);

				pgd_k = pgd_offset_k(vaddr);
				p4d_k = p4d_offset(pgd_k, vaddr);
				pud[j] = *pud_offset(p4d_k, vaddr);
			}
		}
136
	}
137
out:
H
Huang, Ying 已提交
138
	__flush_tlb_all();
139 140

	return save_pgd;
H
Huang, Ying 已提交
141 142
}

143
void __init efi_call_phys_epilog(pgd_t *save_pgd)
H
Huang, Ying 已提交
144 145 146 147
{
	/*
	 * After the lock is released, the original page table is restored.
	 */
148
	int pgd_idx, i;
149
	int nr_pgds;
150 151 152
	pgd_t *pgd;
	p4d_t *p4d;
	pud_t *pud;
153

154 155 156
	if (!efi_enabled(EFI_OLD_MEMMAP)) {
		write_cr3((unsigned long)save_pgd);
		__flush_tlb_all();
157
		return;
158
	}
159

160 161
	nr_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT) , PGDIR_SIZE);

162 163
	for (pgd_idx = 0; pgd_idx < nr_pgds; pgd_idx++) {
		pgd = pgd_offset_k(pgd_idx * PGDIR_SIZE);
164 165
		set_pgd(pgd_offset_k(pgd_idx * PGDIR_SIZE), save_pgd[pgd_idx]);

166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183
		if (!(pgd_val(*pgd) & _PAGE_PRESENT))
			continue;

		for (i = 0; i < PTRS_PER_P4D; i++) {
			p4d = p4d_offset(pgd,
					 pgd_idx * PGDIR_SIZE + i * P4D_SIZE);

			if (!(p4d_val(*p4d) & _PAGE_PRESENT))
				continue;

			pud = (pud_t *)p4d_page_vaddr(*p4d);
			pud_free(&init_mm, pud);
		}

		p4d = (p4d_t *)pgd_page_vaddr(*pgd);
		p4d_free(&init_mm, p4d);
	}

184
	kfree(save_pgd);
185

H
Huang, Ying 已提交
186
	__flush_tlb_all();
187
	early_code_mapping_set_exec(0);
H
Huang, Ying 已提交
188
}
189

190 191 192 193 194 195 196 197 198 199 200
static pgd_t *efi_pgd;

/*
 * We need our own copy of the higher levels of the page tables
 * because we want to avoid inserting EFI region mappings (EFI_VA_END
 * to EFI_VA_START) into the standard kernel page tables. Everything
 * else can be shared, see efi_sync_low_kernel_mappings().
 */
int __init efi_alloc_page_tables(void)
{
	pgd_t *pgd;
201
	p4d_t *p4d;
202 203 204 205 206 207
	pud_t *pud;
	gfp_t gfp_mask;

	if (efi_enabled(EFI_OLD_MEMMAP))
		return 0;

208
	gfp_mask = GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO;
209 210 211 212 213
	efi_pgd = (pgd_t *)__get_free_page(gfp_mask);
	if (!efi_pgd)
		return -ENOMEM;

	pgd = efi_pgd + pgd_index(EFI_VA_END);
214 215 216 217 218
	p4d = p4d_alloc(&init_mm, pgd, EFI_VA_END);
	if (!p4d) {
		free_page((unsigned long)efi_pgd);
		return -ENOMEM;
	}
219

220
	pud = pud_alloc(&init_mm, p4d, EFI_VA_END);
221
	if (!pud) {
222 223
		if (CONFIG_PGTABLE_LEVELS > 4)
			free_page((unsigned long) pgd_page_vaddr(*pgd));
224 225 226 227 228 229 230
		free_page((unsigned long)efi_pgd);
		return -ENOMEM;
	}

	return 0;
}

231 232 233 234 235
/*
 * Add low kernel mappings for passing arguments to EFI functions.
 */
void efi_sync_low_kernel_mappings(void)
{
236 237
	unsigned num_entries;
	pgd_t *pgd_k, *pgd_efi;
238
	p4d_t *p4d_k, *p4d_efi;
239
	pud_t *pud_k, *pud_efi;
240 241 242 243

	if (efi_enabled(EFI_OLD_MEMMAP))
		return;

244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260
	/*
	 * We can share all PGD entries apart from the one entry that
	 * covers the EFI runtime mapping space.
	 *
	 * Make sure the EFI runtime region mappings are guaranteed to
	 * only span a single PGD entry and that the entry also maps
	 * other important kernel regions.
	 */
	BUILD_BUG_ON(pgd_index(EFI_VA_END) != pgd_index(MODULES_END));
	BUILD_BUG_ON((EFI_VA_START & PGDIR_MASK) !=
			(EFI_VA_END & PGDIR_MASK));

	pgd_efi = efi_pgd + pgd_index(PAGE_OFFSET);
	pgd_k = pgd_offset_k(PAGE_OFFSET);

	num_entries = pgd_index(EFI_VA_END) - pgd_index(PAGE_OFFSET);
	memcpy(pgd_efi, pgd_k, sizeof(pgd_t) * num_entries);
261

262 263 264 265 266 267 268 269 270 271 272 273 274 275 276
	/*
	 * As with PGDs, we share all P4D entries apart from the one entry
	 * that covers the EFI runtime mapping space.
	 */
	BUILD_BUG_ON(p4d_index(EFI_VA_END) != p4d_index(MODULES_END));
	BUILD_BUG_ON((EFI_VA_START & P4D_MASK) != (EFI_VA_END & P4D_MASK));

	pgd_efi = efi_pgd + pgd_index(EFI_VA_END);
	pgd_k = pgd_offset_k(EFI_VA_END);
	p4d_efi = p4d_offset(pgd_efi, 0);
	p4d_k = p4d_offset(pgd_k, 0);

	num_entries = p4d_index(EFI_VA_END);
	memcpy(p4d_efi, p4d_k, sizeof(p4d_t) * num_entries);

277 278 279 280 281 282 283
	/*
	 * We share all the PUD entries apart from those that map the
	 * EFI regions. Copy around them.
	 */
	BUILD_BUG_ON((EFI_VA_START & ~PUD_MASK) != 0);
	BUILD_BUG_ON((EFI_VA_END & ~PUD_MASK) != 0);

284 285
	p4d_efi = p4d_offset(pgd_efi, EFI_VA_END);
	p4d_k = p4d_offset(pgd_k, EFI_VA_END);
286 287
	pud_efi = pud_offset(p4d_efi, 0);
	pud_k = pud_offset(p4d_k, 0);
288 289 290 291

	num_entries = pud_index(EFI_VA_END);
	memcpy(pud_efi, pud_k, sizeof(pud_t) * num_entries);

292 293
	pud_efi = pud_offset(p4d_efi, EFI_VA_START);
	pud_k = pud_offset(p4d_k, EFI_VA_START);
294 295 296

	num_entries = PTRS_PER_PUD - pud_index(EFI_VA_START);
	memcpy(pud_efi, pud_k, sizeof(pud_t) * num_entries);
297 298
}

299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327
/*
 * Wrapper for slow_virt_to_phys() that handles NULL addresses.
 */
static inline phys_addr_t
virt_to_phys_or_null_size(void *va, unsigned long size)
{
	bool bad_size;

	if (!va)
		return 0;

	if (virt_addr_valid(va))
		return virt_to_phys(va);

	/*
	 * A fully aligned variable on the stack is guaranteed not to
	 * cross a page bounary. Try to catch strings on the stack by
	 * checking that 'size' is a power of two.
	 */
	bad_size = size > PAGE_SIZE || !is_power_of_2(size);

	WARN_ON(!IS_ALIGNED((unsigned long)va, size) || bad_size);

	return slow_virt_to_phys(va);
}

#define virt_to_phys_or_null(addr)				\
	virt_to_phys_or_null_size((addr), sizeof(*(addr)))

328
int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
329
{
330
	unsigned long pfn, text;
331
	struct page *page;
332
	unsigned npages;
333 334 335 336 337
	pgd_t *pgd;

	if (efi_enabled(EFI_OLD_MEMMAP))
		return 0;

338 339
	efi_scratch.efi_pgt = (pgd_t *)__pa(efi_pgd);
	pgd = efi_pgd;
340

341 342 343 344 345 346
	/*
	 * It can happen that the physical address of new_memmap lands in memory
	 * which is not mapped in the EFI page table. Therefore we need to go
	 * and ident-map those pages containing the map before calling
	 * phys_efi_set_virtual_address_map().
	 */
347
	pfn = pa_memmap >> PAGE_SHIFT;
348
	if (kernel_map_pages_in_pgd(pgd, pfn, pa_memmap, num_pages, _PAGE_NX | _PAGE_RW)) {
349 350 351 352 353 354
		pr_err("Error ident-mapping new memmap (0x%lx)!\n", pa_memmap);
		return 1;
	}

	efi_scratch.use_pgd = true;

355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370
	/*
	 * Certain firmware versions are way too sentimential and still believe
	 * they are exclusive and unquestionable owners of the first physical page,
	 * even though they explicitly mark it as EFI_CONVENTIONAL_MEMORY
	 * (but then write-access it later during SetVirtualAddressMap()).
	 *
	 * Create a 1:1 mapping for this page, to avoid triple faults during early
	 * boot with such firmware. We are free to hand this page to the BIOS,
	 * as trim_bios_range() will reserve the first page and isolate it away
	 * from memory allocators anyway.
	 */
	if (kernel_map_pages_in_pgd(pgd, 0x0, 0x0, 1, _PAGE_RW)) {
		pr_err("Failed to create 1:1 mapping for the first page!\n");
		return 1;
	}

371 372 373 374 375 376
	/*
	 * When making calls to the firmware everything needs to be 1:1
	 * mapped and addressable with 32-bit pointers. Map the kernel
	 * text and allocate a new stack because we can't rely on the
	 * stack pointer being < 4GB.
	 */
377
	if (!IS_ENABLED(CONFIG_EFI_MIXED) || efi_is_native())
378
		return 0;
379 380 381 382 383 384 385 386

	page = alloc_page(GFP_KERNEL|__GFP_DMA32);
	if (!page)
		panic("Unable to allocate EFI runtime stack < 4GB\n");

	efi_scratch.phys_stack = virt_to_phys(page_address(page));
	efi_scratch.phys_stack += PAGE_SIZE; /* stack grows down */

387
	npages = (_etext - _text) >> PAGE_SHIFT;
388
	text = __pa(_text);
389
	pfn = text >> PAGE_SHIFT;
390

391
	if (kernel_map_pages_in_pgd(pgd, pfn, text, npages, _PAGE_RW)) {
392
		pr_err("Failed to map kernel text 1:1\n");
393
		return 1;
394
	}
395 396 397 398

	return 0;
}

399 400
static void __init __map_region(efi_memory_desc_t *md, u64 va)
{
401
	unsigned long flags = _PAGE_RW;
402
	unsigned long pfn;
403
	pgd_t *pgd = efi_pgd;
404 405

	if (!(md->attribute & EFI_MEMORY_WB))
406
		flags |= _PAGE_PCD;
407

408 409
	pfn = md->phys_addr >> PAGE_SHIFT;
	if (kernel_map_pages_in_pgd(pgd, pfn, va, md->num_pages, flags))
410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428
		pr_warn("Error mapping PA 0x%llx -> VA 0x%llx!\n",
			   md->phys_addr, va);
}

void __init efi_map_region(efi_memory_desc_t *md)
{
	unsigned long size = md->num_pages << PAGE_SHIFT;
	u64 pa = md->phys_addr;

	if (efi_enabled(EFI_OLD_MEMMAP))
		return old_map_region(md);

	/*
	 * Make sure the 1:1 mappings are present as a catch-all for b0rked
	 * firmware which doesn't update all internal pointers after switching
	 * to virtual mode and would otherwise crap on us.
	 */
	__map_region(md, md->phys_addr);

429 430 431 432 433 434 435 436 437 438
	/*
	 * Enforce the 1:1 mapping as the default virtual address when
	 * booting in EFI mixed mode, because even though we may be
	 * running a 64-bit kernel, the firmware may only be 32-bit.
	 */
	if (!efi_is_native () && IS_ENABLED(CONFIG_EFI_MIXED)) {
		md->virt_addr = md->phys_addr;
		return;
	}

439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464
	efi_va -= size;

	/* Is PA 2M-aligned? */
	if (!(pa & (PMD_SIZE - 1))) {
		efi_va &= PMD_MASK;
	} else {
		u64 pa_offset = pa & (PMD_SIZE - 1);
		u64 prev_va = efi_va;

		/* get us the same offset within this 2M page */
		efi_va = (efi_va & PMD_MASK) + pa_offset;

		if (efi_va > prev_va)
			efi_va -= PMD_SIZE;
	}

	if (efi_va < EFI_VA_END) {
		pr_warn(FW_WARN "VA address range overflow!\n");
		return;
	}

	/* Do the VA map */
	__map_region(md, efi_va);
	md->virt_addr = efi_va;
}

465 466 467 468 469 470 471
/*
 * kexec kernel will use efi_map_region_fixed to map efi runtime memory ranges.
 * md->virt_addr is the original virtual address which had been mapped in kexec
 * 1st kernel.
 */
void __init efi_map_region_fixed(efi_memory_desc_t *md)
{
472
	__map_region(md, md->phys_addr);
473 474 475
	__map_region(md, md->virt_addr);
}

476
void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size,
477
				 u32 type, u64 attribute)
478 479 480 481 482 483 484 485 486
{
	unsigned long last_map_pfn;

	if (type == EFI_MEMORY_MAPPED_IO)
		return ioremap(phys_addr, size);

	last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size);
	if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) {
		unsigned long top = last_map_pfn << PAGE_SHIFT;
487
		efi_ioremap(top, size - (top - phys_addr), type, attribute);
488 489
	}

490 491 492
	if (!(attribute & EFI_MEMORY_WB))
		efi_memory_uc((u64)(unsigned long)__va(phys_addr), size);

493 494
	return (void __iomem *)__va(phys_addr);
}
495 496 497 498 499

void __init parse_efi_setup(u64 phys_addr, u32 data_len)
{
	efi_setup = phys_addr + sizeof(struct setup_data);
}
B
Borislav Petkov 已提交
500

501
static int __init efi_update_mappings(efi_memory_desc_t *md, unsigned long pf)
B
Borislav Petkov 已提交
502
{
503 504
	unsigned long pfn;
	pgd_t *pgd = efi_pgd;
505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538
	int err1, err2;

	/* Update the 1:1 mapping */
	pfn = md->phys_addr >> PAGE_SHIFT;
	err1 = kernel_map_pages_in_pgd(pgd, pfn, md->phys_addr, md->num_pages, pf);
	if (err1) {
		pr_err("Error while updating 1:1 mapping PA 0x%llx -> VA 0x%llx!\n",
			   md->phys_addr, md->virt_addr);
	}

	err2 = kernel_map_pages_in_pgd(pgd, pfn, md->virt_addr, md->num_pages, pf);
	if (err2) {
		pr_err("Error while updating VA mapping PA 0x%llx -> VA 0x%llx!\n",
			   md->phys_addr, md->virt_addr);
	}

	return err1 || err2;
}

static int __init efi_update_mem_attr(struct mm_struct *mm, efi_memory_desc_t *md)
{
	unsigned long pf = 0;

	if (md->attribute & EFI_MEMORY_XP)
		pf |= _PAGE_NX;

	if (!(md->attribute & EFI_MEMORY_RO))
		pf |= _PAGE_RW;

	return efi_update_mappings(md, pf);
}

void __init efi_runtime_update_mappings(void)
{
539 540 541 542 543 544 545 546
	efi_memory_desc_t *md;

	if (efi_enabled(EFI_OLD_MEMMAP)) {
		if (__supported_pte_mask & _PAGE_NX)
			runtime_code_page_mkexec();
		return;
	}

547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
	/*
	 * Use the EFI Memory Attribute Table for mapping permissions if it
	 * exists, since it is intended to supersede EFI_PROPERTIES_TABLE.
	 */
	if (efi_enabled(EFI_MEM_ATTR)) {
		efi_memattr_apply_permissions(NULL, efi_update_mem_attr);
		return;
	}

	/*
	 * EFI_MEMORY_ATTRIBUTES_TABLE is intended to replace
	 * EFI_PROPERTIES_TABLE. So, use EFI_PROPERTIES_TABLE to update
	 * permissions only if EFI_MEMORY_ATTRIBUTES_TABLE is not
	 * published by the firmware. Even if we find a buggy implementation of
	 * EFI_MEMORY_ATTRIBUTES_TABLE, don't fall back to
	 * EFI_PROPERTIES_TABLE, because of the same reason.
	 */

565
	if (!efi_enabled(EFI_NX_PE_DATA))
B
Borislav Petkov 已提交
566 567
		return;

568
	for_each_efi_memory_desc(md) {
569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584
		unsigned long pf = 0;

		if (!(md->attribute & EFI_MEMORY_RUNTIME))
			continue;

		if (!(md->attribute & EFI_MEMORY_WB))
			pf |= _PAGE_PCD;

		if ((md->attribute & EFI_MEMORY_XP) ||
			(md->type == EFI_RUNTIME_SERVICES_DATA))
			pf |= _PAGE_NX;

		if (!(md->attribute & EFI_MEMORY_RO) &&
			(md->type != EFI_RUNTIME_SERVICES_CODE))
			pf |= _PAGE_RW;

585
		efi_update_mappings(md, pf);
586
	}
B
Borislav Petkov 已提交
587
}
588 589 590 591

void __init efi_dump_pagetable(void)
{
#ifdef CONFIG_EFI_PGT_DUMP
592
	ptdump_walk_pgd_level(NULL, efi_pgd);
593 594
#endif
}
595

596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619
#ifdef CONFIG_EFI_MIXED
extern efi_status_t efi64_thunk(u32, ...);

#define runtime_service32(func)						 \
({									 \
	u32 table = (u32)(unsigned long)efi.systab;			 \
	u32 *rt, *___f;							 \
									 \
	rt = (u32 *)(table + offsetof(efi_system_table_32_t, runtime));	 \
	___f = (u32 *)(*rt + offsetof(efi_runtime_services_32_t, func)); \
	*___f;								 \
})

/*
 * Switch to the EFI page tables early so that we can access the 1:1
 * runtime services mappings which are not mapped in any other page
 * tables. This function must be called before runtime_service32().
 *
 * Also, disable interrupts because the IDT points to 64-bit handlers,
 * which aren't going to function correctly when we switch to 32-bit.
 */
#define efi_thunk(f, ...)						\
({									\
	efi_status_t __s;						\
620 621
	unsigned long __flags;						\
	u32 __func;							\
622
									\
623 624
	local_irq_save(__flags);					\
	arch_efi_call_virt_setup();					\
625
									\
626 627
	__func = runtime_service32(f);					\
	__s = efi64_thunk(__func, __VA_ARGS__);				\
628
									\
629 630
	arch_efi_call_virt_teardown();					\
	local_irq_restore(__flags);					\
631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670
									\
	__s;								\
})

efi_status_t efi_thunk_set_virtual_address_map(
	void *phys_set_virtual_address_map,
	unsigned long memory_map_size,
	unsigned long descriptor_size,
	u32 descriptor_version,
	efi_memory_desc_t *virtual_map)
{
	efi_status_t status;
	unsigned long flags;
	u32 func;

	efi_sync_low_kernel_mappings();
	local_irq_save(flags);

	efi_scratch.prev_cr3 = read_cr3();
	write_cr3((unsigned long)efi_scratch.efi_pgt);
	__flush_tlb_all();

	func = (u32)(unsigned long)phys_set_virtual_address_map;
	status = efi64_thunk(func, memory_map_size, descriptor_size,
			     descriptor_version, virtual_map);

	write_cr3(efi_scratch.prev_cr3);
	__flush_tlb_all();
	local_irq_restore(flags);

	return status;
}

static efi_status_t efi_thunk_get_time(efi_time_t *tm, efi_time_cap_t *tc)
{
	efi_status_t status;
	u32 phys_tm, phys_tc;

	spin_lock(&rtc_lock);

671 672
	phys_tm = virt_to_phys_or_null(tm);
	phys_tc = virt_to_phys_or_null(tc);
673 674 675 676 677 678 679 680 681 682 683 684 685 686 687

	status = efi_thunk(get_time, phys_tm, phys_tc);

	spin_unlock(&rtc_lock);

	return status;
}

static efi_status_t efi_thunk_set_time(efi_time_t *tm)
{
	efi_status_t status;
	u32 phys_tm;

	spin_lock(&rtc_lock);

688
	phys_tm = virt_to_phys_or_null(tm);
689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705

	status = efi_thunk(set_time, phys_tm);

	spin_unlock(&rtc_lock);

	return status;
}

static efi_status_t
efi_thunk_get_wakeup_time(efi_bool_t *enabled, efi_bool_t *pending,
			  efi_time_t *tm)
{
	efi_status_t status;
	u32 phys_enabled, phys_pending, phys_tm;

	spin_lock(&rtc_lock);

706 707 708
	phys_enabled = virt_to_phys_or_null(enabled);
	phys_pending = virt_to_phys_or_null(pending);
	phys_tm = virt_to_phys_or_null(tm);
709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725

	status = efi_thunk(get_wakeup_time, phys_enabled,
			     phys_pending, phys_tm);

	spin_unlock(&rtc_lock);

	return status;
}

static efi_status_t
efi_thunk_set_wakeup_time(efi_bool_t enabled, efi_time_t *tm)
{
	efi_status_t status;
	u32 phys_tm;

	spin_lock(&rtc_lock);

726
	phys_tm = virt_to_phys_or_null(tm);
727 728 729 730 731 732 733 734

	status = efi_thunk(set_wakeup_time, enabled, phys_tm);

	spin_unlock(&rtc_lock);

	return status;
}

735 736 737 738
static unsigned long efi_name_size(efi_char16_t *name)
{
	return ucs2_strsize(name, EFI_VAR_NAME_LEN) + 1;
}
739 740 741 742 743 744 745 746 747

static efi_status_t
efi_thunk_get_variable(efi_char16_t *name, efi_guid_t *vendor,
		       u32 *attr, unsigned long *data_size, void *data)
{
	efi_status_t status;
	u32 phys_name, phys_vendor, phys_attr;
	u32 phys_data_size, phys_data;

748 749 750 751 752
	phys_data_size = virt_to_phys_or_null(data_size);
	phys_vendor = virt_to_phys_or_null(vendor);
	phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
	phys_attr = virt_to_phys_or_null(attr);
	phys_data = virt_to_phys_or_null_size(data, *data_size);
753 754 755 756 757 758 759 760 761 762 763 764 765 766

	status = efi_thunk(get_variable, phys_name, phys_vendor,
			   phys_attr, phys_data_size, phys_data);

	return status;
}

static efi_status_t
efi_thunk_set_variable(efi_char16_t *name, efi_guid_t *vendor,
		       u32 attr, unsigned long data_size, void *data)
{
	u32 phys_name, phys_vendor, phys_data;
	efi_status_t status;

767 768 769
	phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
	phys_vendor = virt_to_phys_or_null(vendor);
	phys_data = virt_to_phys_or_null_size(data, data_size);
770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785

	/* If data_size is > sizeof(u32) we've got problems */
	status = efi_thunk(set_variable, phys_name, phys_vendor,
			   attr, data_size, phys_data);

	return status;
}

static efi_status_t
efi_thunk_get_next_variable(unsigned long *name_size,
			    efi_char16_t *name,
			    efi_guid_t *vendor)
{
	efi_status_t status;
	u32 phys_name_size, phys_name, phys_vendor;

786 787 788
	phys_name_size = virt_to_phys_or_null(name_size);
	phys_vendor = virt_to_phys_or_null(vendor);
	phys_name = virt_to_phys_or_null_size(name, *name_size);
789 790 791 792 793 794 795 796 797 798 799 800 801

	status = efi_thunk(get_next_variable, phys_name_size,
			   phys_name, phys_vendor);

	return status;
}

static efi_status_t
efi_thunk_get_next_high_mono_count(u32 *count)
{
	efi_status_t status;
	u32 phys_count;

802
	phys_count = virt_to_phys_or_null(count);
803 804 805 806 807 808 809 810 811 812 813
	status = efi_thunk(get_next_high_mono_count, phys_count);

	return status;
}

static void
efi_thunk_reset_system(int reset_type, efi_status_t status,
		       unsigned long data_size, efi_char16_t *data)
{
	u32 phys_data;

814
	phys_data = virt_to_phys_or_null_size(data, data_size);
815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841

	efi_thunk(reset_system, reset_type, status, data_size, phys_data);
}

static efi_status_t
efi_thunk_update_capsule(efi_capsule_header_t **capsules,
			 unsigned long count, unsigned long sg_list)
{
	/*
	 * To properly support this function we would need to repackage
	 * 'capsules' because the firmware doesn't understand 64-bit
	 * pointers.
	 */
	return EFI_UNSUPPORTED;
}

static efi_status_t
efi_thunk_query_variable_info(u32 attr, u64 *storage_space,
			      u64 *remaining_space,
			      u64 *max_variable_size)
{
	efi_status_t status;
	u32 phys_storage, phys_remaining, phys_max;

	if (efi.runtime_version < EFI_2_00_SYSTEM_TABLE_REVISION)
		return EFI_UNSUPPORTED;

842 843 844
	phys_storage = virt_to_phys_or_null(storage_space);
	phys_remaining = virt_to_phys_or_null(remaining_space);
	phys_max = virt_to_phys_or_null(max_variable_size);
845

846
	status = efi_thunk(query_variable_info, attr, phys_storage,
847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880
			   phys_remaining, phys_max);

	return status;
}

static efi_status_t
efi_thunk_query_capsule_caps(efi_capsule_header_t **capsules,
			     unsigned long count, u64 *max_size,
			     int *reset_type)
{
	/*
	 * To properly support this function we would need to repackage
	 * 'capsules' because the firmware doesn't understand 64-bit
	 * pointers.
	 */
	return EFI_UNSUPPORTED;
}

void efi_thunk_runtime_setup(void)
{
	efi.get_time = efi_thunk_get_time;
	efi.set_time = efi_thunk_set_time;
	efi.get_wakeup_time = efi_thunk_get_wakeup_time;
	efi.set_wakeup_time = efi_thunk_set_wakeup_time;
	efi.get_variable = efi_thunk_get_variable;
	efi.get_next_variable = efi_thunk_get_next_variable;
	efi.set_variable = efi_thunk_set_variable;
	efi.get_next_high_mono_count = efi_thunk_get_next_high_mono_count;
	efi.reset_system = efi_thunk_reset_system;
	efi.query_variable_info = efi_thunk_query_variable_info;
	efi.update_capsule = efi_thunk_update_capsule;
	efi.query_capsule_caps = efi_thunk_query_capsule_caps;
}
#endif /* CONFIG_EFI_MIXED */